Daily numbers, graphs, analysis thread 2

[Barracker]

A follow on thread from here

Please try to keep it data driven, factual and civil.

www.worldometers.info/coronavirus/country/uk/

Nothing intelligent to add, just and thank you to those who share the figures, data and graphs here each day, I'm finding it so helpful to try and keep a grip on what's going on.

BCF don't encourage mine - I've chucked his dressing gown away

This explains the gender difference

podcasts.apple.com/gb/podcast/more-or-less-behind-the-stats/id267300884

Can I ask a possibly silly question or 2? Today's numbers show they have carried out just over 10k tests (on 7.5k people). Around 40% of those tests were positive. If they are only testing people who have been hospitalised with serious symptoms, I would have expected the rate to be much higher. Have I missed something that says there are lots of false negatives? Also why are they testing some people more than once in a day?

So if they are now at their peak as has been suggested a few times recently then that is 3 weeks from the initial lockdown, which is what was expected?

It's pretty much what I expected, FWIW.

Sky news just gave an embarrassing non-explanation for the daily figures discrepancy which made no sense at all. The figure of 569 they are all quoting cannot possibly be correct, having quoted all four nations and their tallies which add to 636.

What on earth is going on?

@TiddlesUpATree

@murtaman suggests

"Retesting is required for a number of reasons: rechecking status of infected patients, rechecking medical staff/at-risk people; retesting of potentially inaccurate results"

I guess it's not necessarily the same person who is getting retested in the same day .

This explains the gender difference

Gender or sex? The BMJ article above seems to think it’s a bit of both.

(Can’t open the link, I don’t have the app on this device)

The department of health and social care have tweeted 569 (total figure of 2921) so hoping the error is in one of the individual country reports.

oh hang on

they have carried out just over 10k tests (on 7.5k people0

scrub my last sentence from my last post then

Daily mail suggests that some of the Scotland, Wales and NI figures are for after 5pm yesterday - these are not included in the 569 but will be included in tomorrow's figures.

@peridito
This is what Twitter shows.

The BBC website also says this below of Scotland:

*"From next week, the daily figures will also start to include all deaths where the virus is officially "suspected" to have been a factor, even if it has not been confirmed by testing.

Ms Sturgeon said this would increase the number of deaths on record, but would not speculate about how significant this could be.

She said she was "determined that information continues to be as accurate, comprehensive and up to date as possible" and that her government will be "as transparent as we can be"."*

So this may affect the trend for a week - the increase will presumably be higher than otherwise. Unless they announce the figures separately, so that like can be compared with like. (I know there is a non like for like element anyway, but this sounds as though it may mean that there is a significant one-off change.)

Livin- Italy isn't at peak yet.
3-10th April is the estimate.
You are right with the lockdown dates. Nationally 11/3. Lombardia 8/3. Hotspots late Feb.
Lockdown on the 11/3 was pretty similar in restrictions to the current UK one but has been tightened 4 times since then.

I keep posting on the numerous threads about Italy not yet flattening the curve but they seem to be posted by people a) thinking the UK lockdown will be over by Saturday b) people who can't grasp that Italy's lockdown is way tighter than the UK's, c) Italy's majority of infections was very regionalised and there are still large areas untouched d) keep being plopped by the Covid deniers who say it's all scaremongering.

The fact they are still doing that makes me wonder whether people can really be that thick as to not get it, or, alternatively, are Boris plants dressed as Pollyanna.

Livin- Italy isn't at peak yet. 3-10th April is the estimate.

The next week is going to be so crucial. Let's hope Italy does peak as predicted and things start to get better.

I am not sure if it was posted here yet but yesterday’s Eco di Bergamo newspaper did analysis of the deaths recorded in each municipality in the area for March last year and compared it to recorded deaths in the same area for March this year.
From this they think the Covid deaths in the area are more than twice the officially reported number but that these are people who died at home or elsewhere without having been tested.

twitter.com/mickthehack/status/1245131106682392576?s=21

The ploppers are weird aren't they. I really feel like there is some political misinformation agenda going on - god knows why at a time like this

Livin- Italy isn't at peak yet.
3-10th April is the estimate.

What's the basis for this?

The largest number of daily new cases for Italy was over a week ago on the March 21. The largest number of daily deaths was on Monday.

I'm not claiming either of those dates for the peak necessarily (you'd need to fit a curve to work that out), but for new cases (and probably for new deaths) Italy seems to be flatlining now.

Nicola Sturgeon made it clear in her briefing the Scotland's daily total of deaths of 50 was split - 10 being the overnight figures from individual health boards following confirmed tests and 40 from other sources where COVID-19 was referenced.

All 40 of those could not be added to today's figure and would after further analysis be added to the relevant dates.

"From next week, the daily figures will also start to include all deaths where the virus is officially "suspected" to have been a factor, even if it has not been confirmed by testing.

Unless the past figures can be similarly re-based, I'm not a fan of this. What I want to see is a consistent methodology to generate the numbers. Carry out an additional step to generate separate refined numbers for those purposes where the absolute number is important rather than the trend.

See below @nauticant - think your concerns have been addressed?

Gender or sex? The BMJ article above seems to think it’s a bit of both.

I think it's quite likely to be. Especially in China where many more men smoke etc. But it seems fairly clear there is some sex linked difference.

So long as there can be continuity with a number series having been generated on the same basis then that' looks sensible PinkGinny.

thecanterbury I can assure you I am neither of those things. I was just mulling it over as I thought I had heard that Italy was at peak. If I'm wrong, I'm wrong.

And for my second question of the day @QuentinWinters - what's a plopper?

A plopper is someone who drops goady posts into a thread which stir up all kinds of annoyance and then doesn't engage with any of the kerfuffle they've caused. Well, except to cause more. At which point they get promoted to troll.